Paper Citation Record · LEDGER

Why Low-Precision Transformer Training Fails: An Analysis on Flash Attention

As of 6 August 2026, this Paper Citation Record lists 35 of 35 outbound references and 5 inbound Pith citation observations for arXiv:2510.04212.

A citation records a reference. It does not transfer a finding from one paper to another.

pith.paper-citation-record.v1
2510.04212 v3

Coverage vector

measured 35 of 35 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-05-18T10:01:56.131253Z

measured 40 of 40 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-06T06:34:29.942622+00:00

measured 5 of 5 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-08-04T15:20:16.164286Z

measured 0 of 1 external citation measurements

A source-named dated measurement, never combined with another source.

Source: cited_works

Reference resolution

35 of 35 outbound references displayed

verified exact23
verified fuzzy4
unresolved0
parse uncertain0
malformed identifier2
metadata mismatch6

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 84ef2b0d-6d00-4360-8707-47ea35c43375 · outbound

This paper cites Scalify: scale propagation for efficient low-precision LLM training.

Why Low-Precision Transformer Training Fails: An Analysis on Flash Attention Scalify: scale propagation for efficient low-precision LLM training

Reference 1

Resolution: verified exact; arxiv_id, observed 2026-05-18T10:02:31.677451Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation c8ce8386-332b-4824-a827-d4101c1410d5 · outbound

This paper cites u-$\mu$P: The Unit-Scaled Maximal Update Parametrization.

Why Low-Precision Transformer Training Fails: An Analysis on Flash Attention u-$\mu$P: The Unit-Scaled Maximal Update Parametrization

Reference 2

Resolution: verified exact; arxiv_id, observed 2026-05-18T10:02:31.733884Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation acad3c74-c1c2-4dc6-bcbf-cf9f912f814c · outbound

This paper cites Language models are few-shot learners.Advances in neural information processing systems, 33:1877–1901.

Why Low-Precision Transformer Training Fails: An Analysis on Flash Attention Language models are few-shot learners.Advances in neural information processing systems, 33:1877–1901

Reference 3

Resolution: verified fuzzy; raw_fallback, observed 2026-05-18T10:02:31.810103Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 2747e26a-663e-407e-9239-46432204b867 · outbound

This paper cites Scaling FP8 training to trillion-token LLMs.

Why Low-Precision Transformer Training Fails: An Analysis on Flash Attention Scaling FP8 training to trillion-token LLMs

Reference 4

Resolution: verified exact; arxiv_id, observed 2026-05-18T10:02:31.710568Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation a45bdfba-62a7-40cd-8e79-c5196ab742e5 · outbound

This paper cites Aaron Gokaslan, Vanya Cohen, Ellie Pavlick, and Stefanie Tellex.

Why Low-Precision Transformer Training Fails: An Analysis on Flash Attention Aaron Gokaslan, Vanya Cohen, Ellie Pavlick, and Stefanie Tellex

Reference 5

Resolution: verified fuzzy; raw_fallback, observed 2026-05-18T10:02:31.806824Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 0a266c88-bd65-4dac-bdbe-67731ecbcadc · outbound

This paper cites Is Flash Attention Stable?.

Why Low-Precision Transformer Training Fails: An Analysis on Flash Attention Is Flash Attention Stable?

Reference 6

Resolution: metadata mismatch; arxiv_id, observed 2026-05-18T10:02:31.719792Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 51ef9cb5-1de1-4065-b39d-0102ff8bb07f · outbound

This paper cites Low-Precision Training of Large Language Models: Methods, Challenges, and Opportunities.

Why Low-Precision Transformer Training Fails: An Analysis on Flash Attention Low-Precision Training of Large Language Models: Methods, Challenges, and Opportunities

Reference 7

Resolution: verified exact; arxiv_id, observed 2026-07-30T01:18:51.459998Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation cc065a55-4627-413b-9d94-158782c9e583 · outbound

This paper cites Query-Key Normalization for Transformers.

Why Low-Precision Transformer Training Fails: An Analysis on Flash Attention Query-Key Normalization for Transformers

Reference 8

Resolution: verified exact; arxiv_id, observed 2026-05-18T10:02:31.673927Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 8f6b5e54-03fc-4290-a857-7034d2da5616 · outbound

This paper cites Training Compute-Optimal Large Language Models.

Why Low-Precision Transformer Training Fails: An Analysis on Flash Attention Training Compute-Optimal Large Language Models

Reference 9

Resolution: verified exact; local_arxiv, observed 2026-05-18T10:02:31.752653Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 88aff18f-3fb0-4d42-b513-6972e031ff0f · outbound

This paper cites SPAM: Spike-Aware Adam with Momentum Reset for Stable LLM Training.

Why Low-Precision Transformer Training Fails: An Analysis on Flash Attention SPAM: Spike-Aware Adam with Momentum Reset for Stable LLM Training

Reference 10

Resolution: verified exact; arxiv_id, observed 2026-05-18T10:02:31.706582Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 52e00fa9-ec25-4b75-8a7a-7be7ccc302f3 · outbound

This paper cites A Study of BFLOAT16 for Deep Learning Training.

Why Low-Precision Transformer Training Fails: An Analysis on Flash Attention A Study of BFLOAT16 for Deep Learning Training

Reference 11

Resolution: verified exact; local_arxiv, observed 2026-05-18T10:02:31.738309Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 53f66d0e-6c7b-41f6-96be-f52c7003805a · outbound

This paper cites Kimi K2: Open Agentic Intelligence.

Why Low-Precision Transformer Training Fails: An Analysis on Flash Attention Kimi K2: Open Agentic Intelligence

Reference 12

Resolution: verified exact; local_arxiv, observed 2026-05-18T10:02:31.756129Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 0d9e66a6-2abb-4607-b5bf-4c19b90c21b2 · outbound

This paper cites DeepSeek-V3 Technical Report.

Why Low-Precision Transformer Training Fails: An Analysis on Flash Attention DeepSeek-V3 Technical Report

Reference 13

Resolution: verified exact; local_arxiv, observed 2026-05-18T10:02:31.702310Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 3e77396a-2006-4532-94df-1611f448be03 · outbound

This paper cites Mixed Precision Training With 8-bit Floating Point.

Why Low-Precision Transformer Training Fails: An Analysis on Flash Attention Mixed Precision Training With 8-bit Floating Point

Reference 14

Resolution: verified exact; local_arxiv, observed 2026-05-18T10:02:31.684327Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 7b607e40-1368-4393-880c-4fb464af087f · outbound

This paper cites Mixed Precision Training.

Why Low-Precision Transformer Training Fails: An Analysis on Flash Attention Mixed Precision Training

Reference 15

Resolution: verified exact; local_arxiv, observed 2026-05-18T10:02:31.654634Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 267a6714-9fda-4947-8fd6-a76fb131a427 · outbound

This paper cites FP8 Formats for Deep Learning.

Why Low-Precision Transformer Training Fails: An Analysis on Flash Attention FP8 Formats for Deep Learning

Reference 16

Resolution: verified exact; local_arxiv, observed 2026-05-18T10:02:31.723095Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 89f76f7b-7d55-4d47-bbd3-2d7e6cf46b53 · outbound

This paper cites A Theory on Adam Instability in Large-Scale Machine Learning.

Why Low-Precision Transformer Training Fails: An Analysis on Flash Attention A Theory on Adam Instability in Large-Scale Machine Learning

Reference 17

Resolution: metadata mismatch; arxiv_id, observed 2026-05-18T10:02:31.691756Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 1e1dbe79-4ae7-41dc-939e-c2a6900e0c92 · outbound

This paper cites nanoGPT Issue.

Why Low-Precision Transformer Training Fails: An Analysis on Flash Attention nanoGPT Issue

Reference 19

Resolution: verified fuzzy; raw_fallback, observed 2026-05-18T10:02:31.803305Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 93c4ca0f-0f0f-492e-9f4b-9839f0659c4e · outbound

This paper cites 8-bit Numerical Formats for Deep Neural Networks.

Why Low-Precision Transformer Training Fails: An Analysis on Flash Attention 8-bit Numerical Formats for Deep Neural Networks

Reference 20

Resolution: verified exact; arxiv_id, observed 2026-05-18T10:02:31.639481Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 65c71bf6-d9cd-4002-9fb2-6a0f447ec470 · outbound

This paper cites FP8-LM: Training FP8 Large Language Models.

Why Low-Precision Transformer Training Fails: An Analysis on Flash Attention FP8-LM: Training FP8 Large Language Models

Reference 21

Resolution: metadata mismatch; arxiv_id, observed 2026-05-18T10:02:31.727622Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 56d5338f-e883-4b21-a256-22388e7fa288 · outbound

This paper cites Training and inference of large language models using 8-bit floating point.

Why Low-Precision Transformer Training Fails: An Analysis on Flash Attention Training and inference of large language models using 8-bit floating point

Reference 22

Resolution: verified exact; arxiv_id, observed 2026-05-18T10:02:31.696207Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation bf83668c-6f1c-4e1e-b0be-bd887d7e2282 · outbound

This paper cites Gated Attention for Large Language Models: Non-linearity, Sparsity, and Attention-Sink-Free.

Why Low-Precision Transformer Training Fails: An Analysis on Flash Attention Gated Attention for Large Language Models: Non-linearity, Sparsity, and Attention-Sink-Free

Reference 23

Resolution: verified exact; local_arxiv, observed 2026-05-18T10:02:31.643403Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 35f406cc-da44-44be-9695-590dda5c8203 · outbound

This paper cites Qwen3 Technical Report.

Why Low-Precision Transformer Training Fails: An Analysis on Flash Attention Qwen3 Technical Report

Reference 24

Resolution: metadata mismatch; local_arxiv, observed 2026-05-18T10:02:31.647731Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation f223dff7-69e1-434c-8aa5-d4a7d67c0139 · outbound

This paper cites Scaling Language Models: Methods, Analysis & Insights from Training Gopher.

Why Low-Precision Transformer Training Fails: An Analysis on Flash Attention Scaling Language Models: Methods, Analysis & Insights from Training Gopher

Reference 25

Resolution: verified exact; local_arxiv, observed 2026-05-18T10:02:31.746740Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 22a56c48-f98b-43e8-84e5-c06205cf2fcc · outbound

This paper cites Methods of improving LLM training stability.

Why Low-Precision Transformer Training Fails: An Analysis on Flash Attention Methods of improving LLM training stability

Reference 26

Resolution: verified exact; arxiv_id, observed 2026-05-18T10:02:31.742547Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation ab409d59-1954-493a-bdfc-a17037221042 · outbound

This paper cites LLaMA: Open and Efficient Foundation Language Models.

Why Low-Precision Transformer Training Fails: An Analysis on Flash Attention LLaMA: Open and Efficient Foundation Language Models

Reference 27

Resolution: verified exact; local_arxiv, observed 2026-05-18T10:02:31.651207Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 0ee60707-4b4f-478a-ab91-f55567e38442 · outbound

This paper cites Training LLMs with MXFP4.

Why Low-Precision Transformer Training Fails: An Analysis on Flash Attention Training LLMs with MXFP4

Reference 28

Resolution: verified exact; arxiv_id, observed 2026-05-18T10:02:31.760017Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 7eee6e6f-7efe-42c7-a26d-c059d5f17b33 · outbound

This paper cites Optimizing Large Language Model Training Using FP4 Quantization.

Why Low-Precision Transformer Training Fails: An Analysis on Flash Attention Optimizing Large Language Model Training Using FP4 Quantization

Reference 29

Resolution: verified exact; arxiv_id, observed 2026-05-20T00:00:17.330307Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 8f603f96-661d-4b81-a8c9-86a00ab49069 · outbound

This paper cites Mitchell Wortsman, Tim Dettmers, Luke Zettlemoyer, Ari S.

Why Low-Precision Transformer Training Fails: An Analysis on Flash Attention Mitchell Wortsman, Tim Dettmers, Luke Zettlemoyer, Ari S

Reference 30

Resolution: verified fuzzy; raw_fallback, observed 2026-05-18T10:02:31.799841Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 7b776de5-23a8-4a73-bd64-03fed27455ee · outbound

This paper cites Efficient Streaming Language Models with Attention Sinks.

Why Low-Precision Transformer Training Fails: An Analysis on Flash Attention Efficient Streaming Language Models with Attention Sinks

Reference 31

Resolution: metadata mismatch; local_arxiv, observed 2026-05-18T10:02:31.687763Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation ab5518ee-fc94-4220-a46e-72af56d69016 · outbound

This paper cites Tensor Programs V: Tuning Large Neural Networks via Zero-Shot Hyperparameter Transfer.

Why Low-Precision Transformer Training Fails: An Analysis on Flash Attention Tensor Programs V: Tuning Large Neural Networks via Zero-Shot Hyperparameter Transfer

Reference 32

Resolution: verified exact; arxiv_id, observed 2026-05-18T10:02:31.664367Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 94f1e29d-3a57-407c-8f9e-e46d13956016 · outbound

This paper cites A Spectral Condition for Feature Learning.

Why Low-Precision Transformer Training Fails: An Analysis on Flash Attention A Spectral Condition for Feature Learning

Reference 33

Resolution: metadata mismatch; arxiv_id, observed 2026-05-18T10:02:31.669691Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 2afeead0-f78f-4c56-b46a-2fd1adf104bb · outbound

This paper cites Towards Efficient Pre-training: Exploring FP4 Precision in Large Language Models.

Why Low-Precision Transformer Training Fails: An Analysis on Flash Attention Towards Efficient Pre-training: Exploring FP4 Precision in Large Language Models

Reference 34

Resolution: verified exact; arxiv_id, observed 2026-05-18T10:02:31.635398Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 7d863fef-6e0c-4126-91d7-be66321cb39a · outbound

This paper cites gradient spikes.

Why Low-Precision Transformer Training Fails: An Analysis on Flash Attention gradient spikes

Reference 35

Resolution: malformed identifier; raw_fallback, observed 2026-05-18T10:02:31.796249Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation a5f9a6f2-e73d-488c-b77f-1eed422f3313 · outbound

This paper cites Seg en Ã st 're ich s ho hem S oh ne / Un ser m Kaiser Ferdinand !.

Why Low-Precision Transformer Training Fails: An Analysis on Flash Attention Seg en Ã st 're ich s ho hem S oh ne / Un ser m Kaiser Ferdinand !

Reference 36

Resolution: malformed identifier; raw_fallback, observed 2026-05-18T10:02:31.793196Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Pith citing papers

Observation d8e3f90e-889b-4d2a-94c3-39e22dfc17bc · inbound

Reference Traces for Auditing Invisible Weight Updates and Guiding Exact-Budget Protection cites this paper.

Reference Traces for Auditing Invisible Weight Updates and Guiding Exact-Budget Protection Why Low-Precision Transformer Training Fails: An Analysis on Flash Attention

Reference 13

Resolution: unresolved; no resolver link, observed 2026-07-14T15:32:26.691504Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ab61e964-02b0-4497-874c-767d0fae2e8b · inbound

Reference Traces for Auditing Invisible Weight Updates and Guiding Exact-Budget Protection cites this paper.

Reference Traces for Auditing Invisible Weight Updates and Guiding Exact-Budget Protection Why Low-Precision Transformer Training Fails: An Analysis on Flash Attention

Reference 34

Resolution: unresolved; no resolver link, observed 2026-08-02T07:54:47.068638Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 55e5c0eb-4757-49d1-9eb8-3d238c96c5ab · inbound

An Isotropy-Preserving Spectral Cap for Muon: Theory and Three Case Studies cites this paper.

An Isotropy-Preserving Spectral Cap for Muon: Theory and Three Case Studies Why Low-Precision Transformer Training Fails: An Analysis on Flash Attention

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-01T11:54:35.779079Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f11b71b5-ce65-43a5-9ff8-5b4734344cb7 · inbound

Automated Numerical Stability Analysis of Deep Learning Operators cites this paper.

Automated Numerical Stability Analysis of Deep Learning Operators Why Low-Precision Transformer Training Fails: An Analysis on Flash Attention

Reference 50

Resolution: unresolved; no resolver link, observed 2026-08-01T02:20:39.957037Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 55a7f5a8-c489-48d9-a2be-bfe3e7067fcb · inbound

One QK Channel, Many Sources: Guarding Low-Precision Attention Collapse cites this paper.

One QK Channel, Many Sources: Guarding Low-Precision Attention Collapse Why Low-Precision Transformer Training Fails: An Analysis on Flash Attention

Reference 28

Resolution: unresolved; no resolver link, observed 2026-08-04T15:20:16.164286Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.